GunSpec

Data Quality

Data completeness and confidence metrics. Understand how thoroughly each record is populated and identify low-confidence entries that may need enrichment.

Optional AuthExplorer+

The work generated from what the quality checks find, and where each item has got to.

GET/v1/data/tasks

  • stringheader

    Optional. Without a key this endpoint answers under the public policy; a key raises your limits and unlocks plan-gated fields. Authorization: Bearer <key> is accepted in its place.

  • string

    Only tasks in this state.

    Accepts
  • string

    Only tasks for one catalog.

    Accepts
  • string

    Only one check's tasks, by the id the data quality pages print (firearm-weight-fits-its-size); the quality: prefix is optional.

    Example
  • string

    Only tasks about one record field.

    Example
  • string

    Only tasks about one record.

    Example
  • string

    A search: a task number (12879 or #12879), or text found in the record id or the check id.

    Max length
    120
    Example
  • string

    Only one kind of task.

    Accepts
  • string

    Where the work stands: queued, working, proposed (answered and waiting on a person) or accepted.

    Accepts
  • string

    Several checks' tasks at once, comma-separated (at most 40), for a page listing every check that rests on one method. Each id matches with or without its quality: prefix.

    Example
  • string

    open is every task not yet closed (queued or being worked); closed is every closed task, newest closure first.

    Accepts
  • string

    Closed tasks by who closed them, newest closure first: redback (the platform's agent filed a conclusion nobody has judged yet), redback_accepted (a person accepted what the agent filed) or team (a person worked or closed it, or corrected the record so the check no longer finds it).

    Accepts
  • string

    research adds each task's research inputs: the rule, the steps, the evidence and the record's own fields. Never the proposal.

    Accepts
  • integerdefault50

    How many to return per page.

    Range
    1 – 500
    Example
  • integerdefault1

    The page, from 1.

    Minimum
    1
    Example

  • object[]

    The response payload.

  • integer

    How many tasks match.

  • integer

    This page.

  • integer

    Tasks per page.

  • object[]

    Most recently seen first; with state=closed or closer, most recently closed first.

  • string

    Stable id, derived from the finding rather than assigned, so the same finding is the same task tomorrow.

  • integer | null

    The task's number; /v1/data/tasks/{number} loads it.

  • object

    With include=research only: what a researcher needs, read from the task. Never its result.

  • string

    What the asset is: silhouette, render, photo, schematic, model. /v1/firearms/{id}/media/{selector} addresses an asset by this.

  • string

    Which catalog.

  • string

    The record, or for a vocabulary task the value being decided.

  • string | null

    The check id or field this came from.

  • string

    Production status, one of in_production, discontinued, out_of_production, in_service, limited_production, prototype. Discontinued records are kept deliberately: a rifle built in 2004 wears parts nobody sells today.

  • number

    How many records this would fix. One for most; a vocabulary decision covers every record holding that spelling.

  • string

    When the finding was first seen.

  • string | null

    When a worker began, or null.

  • string | null

    When it closed, or null.

  • number

    Current page number

  • number

    Items per page

  • number

    Total matching records (Builder and above)

  • 400VALIDATION_ERRORValidation error
  • 429RATE_LIMITEDRate limit exceeded
  • 500INTERNAL_ERRORUnexpected server error

Minimum tier
Explorer+
Monthly quota
200/mo
Rate limit
10/min
Daily cap
50

Figures shown are for this endpoint's minimum tier. Higher tiers raise every limit.